Latest

Every published item, in order · 56 items
Oct 22
Oct 20
Oct 17
IndustrySelected05:18
Andrew Ng's The Batch: verification drives AI progress

The verification-as-bottleneck framing offers a practical filter for deciding where to deploy AI today: benchmark gains in code or math may not transfer to open-ended judgment work in your domain. Practitioners building automation roadmaps can use this lens to prioritise workflows with clear correctness checks. The newsletter's roundup of Claude 5.5 releases and Mistral Large 4 also serves as a quick catch-up for readers tracking competing frontier models.

DeepLearning.AI@DeepLearningAISource
Oct 16
Oct 15
Oct 14
ProductsBuildersSelected00:45
Artificial Analysis Launches Decision Index v1.0 Benchmark

Decision models trade some accuracy for far lower cost and latency, making them viable for high-volume routing, moderation and classification workloads where a full language model is overkill. The finding that most models are poorly calibrated on knowledge tasks, yet reasonably calibrated on classification, tells builders where this approach is safe to deploy. Teams weighing Jev 1.13 against Microsoft-Decision-1 or Perplexity's Decider v1.1 get a direct cost-per-task comparison.

官方账号Artificial Analysis@ArtificialAnlysSource
Oct 11
Oct 10
Oct 9
IndustryUS vs ChinaMulti-source00:34
Chinese contentThe Information 节目预告:电网争议、Sriram Krishnan 新基金与 Claude Haiku 5.5

这期聊了三个点:Amazon 的电网危机、Sriram Krishnan 的 5 亿美元新基金,还有 Claude Haiku 5.5 和 Mistral、Reflection 新模型的对比,想了解这几家的可以看。

Event
The Information@theinformation11 sourcesSource
Oct 8
ProductsBuildersMulti-sourceSelected06:05
Simon Willison Tests Claude Haiku 5.5 Pricing With Pelican Benchmark

The pelican test is a quick, repeatable way to sanity-check a small model's instruction following, and pairing it with a token-price breakdown shows how capability per dollar has shifted across the Haiku line. Developers choosing a cheap workhorse model for high-volume tasks can use the comparison to decide whether upgrading from an earlier Haiku pays off. Watch the input/output price ratio, since it matters most for long-context workloads.

Event
官方账号Simon Willison@simonw11 sourcesSource
Sep 17
AITOP16:09
Anthropic拟10月在新加坡设亚太第五个办事处 开...

Anthropic拟10月在新加坡设亚太第五个办事处 开放本地招聘 数据显示,按人口比例计算,新加坡的Claude.ai使用水平在121个国家中排名第二。 我们观察到,人工智能公司Anthropic计划于10月在新加坡设立其亚太地区的第五…

AITOPRead
Aug 13
Aug 7
AITOP17:04
出口结构剧变:中国卖给世界的,不再是产品而是解法

中国出口"新新三样":从打火机到爬墙机器人,世界不再只是买便宜货 我们观察到一组清晰的数据信号:中国出口的"新新三样"已经从过去的服装、家具、家电,换成机器人、AI模型和创新药物。这不是产业升级的修辞游戏,而是发生在澳大利亚和巴西的实证样…

AITOPRead
Aug 2